Abstract
This paper examined the extent to which the use of appropriate selection factors to determine which studies should be reviewed is crucial to the validity of meta-analysis. Results obtained in a recent National Institute of Education investigation of desegregatory effects on blacks' achievement were appraised. Findings indicated that a few “outlier” studies skewed the over-all effect size. The present study employed microanalysis of the most extreme single “outlier” study, which alone accounted for a substantial amount of variance in the meta-analysis of 19 studies. Detailed analysis indicated that the targeted “outlier” study was inappropriate for use in the meta-analysis because it contains a number of methodological and statistical problems.
Get full access to this article
View all access options for this article.